Medical Policy

This document addresses blood-based biomarker tests for multiple sclerosis (MS). Examples are:

• Glial fibrillary acidic protein (GFAP)
• Neurofilament light chain (NfL)
• Octave^® Multiple Sclerosis Disease Activity (MSDA) test (Octave Biosciences, Menlo Park, CA)

Blood-based biomarker tests for MS are blood tests designed to either expedite the diagnosis of MS or as a prognostic tool to measure the risk for rapid progression of disability in individuals with relapsing-remitting MS (RRMS) or clinically isolated syndrome (CIS).

Note: For a high-level overview of this document, please see “Summary for Members and Families” below.

Investigational and Not Medically Necessary:

Blood-based biomarker tests for multiple sclerosis are considered investigational and not medically necessary for all uses.

This document describes clinical studies and expert recommendations, and explains why blood-based biomarker tests for multiple sclerosis (MS) are not clinically appropriate. The following summary does not replace the medical necessity criteria or other information in this document. The summary may not contain all of the relevant criteria or information. This summary is not medical advice. Please check with your healthcare provider for any advice about your health.

Key Information

Blood-based biomarker tests are blood tests that are being studied to help diagnose MS or predict how the disease may change over time. Examples include glial fibrillary acidic protein (GFAP), neurofilament light chain (NfL), and the Octave Multiple Sclerosis Disease Activity (MSDA) Test. Researchers hope these tests may provide information about disease activity along with other tools, such as magnetic resonance imaging (MRI). However, current research has not shown that these blood tests improve diagnosis, guide treatment in a way that improves health, or lead to better long-term outcomes. Therefore, these tests are not considered clinically appropriate for routine use.

What the Studies Show

The Octave MSDA Test has been shown to measure disease activity and may influence treatment decisions. However, studies have not shown that using the test leads to fewer MS relapses, slower disability progression, better physical function, or improved quality of life. Better studies are needed to know if the Octave MSDA Test improves health.

Researchers continue to study blood biomarkers such as NfL and GFAP. Some studies suggest that rising NfL levels may help identify returning disease activity after stopping MS treatment or may help predict which people with clinically isolated syndrome (CIS), a condition considered a possible indicator of possible MS, are more likely to develop MS. GFAP may also provide information about disease activity in some situations. However, these findings came from relatively small or limited studies, and the proposed test results still need to be confirmed in larger groups of people before they can be used routinely.

A recent review found that NfL has the strongest evidence among blood biomarkers currently being studied. Even so, NfL is not specific to MS because levels can also increase in other conditions that affect the nervous system. Researchers also noted that standard testing methods, reference ranges, and guidance on how to use these tests in routine care are still needed.

Expert organizations have reached similar conclusions. The French Multiple Sclerosis Think Tank (Groupe de Réflexion sur la Sclérose en Plaques [GRESEP]), the National Multiple Sclerosis Society (NMSS), the European Committee for Treatment and Research in Multiple Sclerosis (ECTRIMS), and the MS Phenotype Group all conclude that there is not enough evidence to recommend routine blood biomarker testing for diagnosing or monitoring MS.

At this time, no blood biomarker test for MS has received United States Food and Drug Administration (FDA) approval for this use. Although two NfL tests have received FDA Breakthrough Device Designation, this designation does not mean the tests have been approved. The Octave MSDA test is considered laboratory-developed test (LDT), and is allowed on the market in the United States (U.S.) without FDA approval.

Is this Clinically Appropriate?

Blood-based biomarker tests for multiple sclerosis are not clinically appropriate because they have not been proven to improve health. Studies have not shown that using these tests improves diagnosis, changes long-term care in a way that benefits people, or leads to better health outcomes than current standard evaluation with a healthcare provider and MRI.

(Return to Description/Scope)

Summary

Blood-based biomarker tests for multiple sclerosis (MS), including glycan antibody assays, serum neurofilament light chain (sNfL), the Octave Multiple Sclerosis Disease Activity (MSDA) Test, and other emerging serum biomarkers are designed to assist with diagnosis or monitoring of disease activity. The evidence continues to demonstrate that while biomarkers such as sNfL correlate with disease activity, MRI lesion burden, disability progression, and treatment response, they have not been shown to reliably establish the diagnosis of MS or independently improve clinical outcomes when incorporated into routine management. Earlier glycan-based assays have not been consistently validated in larger studies and are no longer commercially available. More recent evidence evaluating the Octave MSDA Test suggests that test results may influence treatment decisions in clinical practice. However, available studies have not demonstrated that biomarker-guided management results in improved clinical outcomes, such as reduced relapse rates, delayed disability progression, or improved quality of life. Additional blood biomarkers, including glial fibrillary acidic protein (GFAP), anti-KIR4.1 antibodies, antiphospholipid antibodies, anti-myelin antibodies, osteopontin, microRNAs, and plasma NfL, remain under evaluation, with insufficient evidence to support routine clinical use. No blood-based biomarker test for MS has received United States Food and Drug Administration (FDA) approval for this indication, although two NfL assays have been granted FDA Breakthrough Device Designation. Professional organizations, including the National Multiple Sclerosis Society (NMSS), the European Committee for Treatment and Research in Multiple Sclerosis (ECTRIMS), and the MS Phenotype Group, acknowledge the potential role of biomarkers but conclude that current evidence is insufficient to incorporate blood biomarkers into routine MS diagnosis or disease classification. Similarly, recommendations from the French Multiple Sclerosis Think Tank (Groupe de Réflexion sur la Sclérose en Plaques [GRESEP]) indicate that serum biomarkers have not been established as useful for the routine diagnosis or differential diagnosis of MS. Overall, the published evidence remains insufficient to demonstrate that blood-based biomarker testing improves diagnostic accuracy, alters long-term clinical management, or improves health outcomes compared with current standard clinical evaluation and magnetic resonance imaging (MRI).

Discussion

Early reports showed potential promise for glycan antibody assays for use as an aid in the diagnosis or prognosis of MS (Brettschneider, 2009; Freedman, 2009; Schwartz, 2006). However, outcomes from these studies have not been confirmed, and large, well-designed trials are warranted to validate findings.

Freedman (2012) conducted a prospective biomarker analysis using serum samples from the BEtaseron^® in Newly Emerging Multiple Sclerosis For Initial Treatment (BENEFIT) trial to evaluate whether the gMS-Classifier 1a panel of four anti-α-glucose IgM antibodies (anti-GAGA2, anti-GAGA3, anti-GAGA4, and anti-GAGA6), could predict disease activity and disability progression in individuals with clinically isolated syndrome (CIS). Of the 468 participants enrolled in BENEFIT, 286 (61%) had sufficient baseline serum available for biomarker analysis. The primary objective was to determine whether a positive gMS-Classifier1 predicted earlier conversion to clinically definite multiple sclerosis (CDMS). The biomarker did not predict time to CDMS or conversion to McDonald 2005 MS, and demonstrated poor diagnostic performance for early conversion, with a positive predictive value of 38.0%, 95% confidence interval (CI) 24.6-51.5%. However, over 5 years of follow-up, participants who were gMS-Classifier1 positive had approximately twice the risk of confirmed disability progression compared with biomarker-negative participants (hazard ratio [HR], 2.05; 95% Confidence Interval [CI], 1.2-3.5; p=0.009). When baseline Expanded Disability Status Scale (EDSS) was included in the model, the association was attenuated and no longer statistically significant (HR, 1.63; 95% CI, 0.95-2.8; p=0.07). Kaplan-Meier estimates showed that by year 5, 41.8% of gMS-Classifier1 positive participants experienced confirmed EDSS progression compared with 23.3% of biomarker-negative participants. The biomarker was not associated with annualized relapse rate (p=0.13) or MRI outcomes. Limitations include that the study evaluated a subset of BENEFIT participants (286 of 468) because serum samples were unavailable for all enrolled individuals, introducing the potential for selection bias. Although the analysis was prospective, it represented a secondary biomarker analysis rather than a trial designed specifically to validate the assay. The study population differed substantially from the earlier cohort in which the biomarker had been developed, including recruitment from 98 centers across Europe and Canada, potentially contributing to reduced predictive performance. Most participants did not convert to CDMS during follow-up, limiting power to detect differences in relapse-related outcomes. Biomarker classification thresholds were adapted from a previous fluorescence assay to an enzyme immunoassay platform, which may have affected reproducibility. Finally, the findings were not independently validated in an external cohort.

The use of sNfL is being investigated as a biomarker of neuronal damage in individuals with MS. It has been proposed that the biomarker be used to monitor disease activity, response to drug therapy and to prognosticate the course of the disease in individuals with MS (Atkas, 2020; Benkert, 2022; Cai, 2018; Calabresi, 2021; Hanninen, 2020; Rosso, 2020).

Seiberl (2023) evaluated sNfL as a biomarker in a prospective, real-world cohort of 14 individuals with relapsing multiple sclerosis treated with cladribine (CLAD). Participants had a mean age of 35.4 years, most were female (86%), and median follow-up was 41.5 months. CLAD was associated with a significant reduction in sNfL from 24.7 pg/mL at baseline to 8.8 pg/mL at 12 months (p=0.0008), representing an approximately 65% decrease. There was a decrease of sNfL in 13 of 14 individuals (93%). NEDA-3 status was maintained in 71% at 12 months, 57% at 24 months, and 36% at 36 months. During follow-up, 29% had relapses, 43% had MRI activity, 36% had EDSS worsening, and 29% switched to another disease-modifying therapy (DMT). However, baseline sNfL, 12-month sNfL, and the sNfL ratio did not correlate with time to loss of NEDA-3, relapse occurrence, MRI activity, EDSS progression, treatment switch, or sustained NEDA-3. Limitations include a small sample size, single-center real-world design, and lack of a control group. The cohort was heterogeneous, with most individuals previously treated with other DMTs and many having recent disease activity before CLAD initiation. sNfL was only measured at baseline and 12 months, so the study could not assess whether more frequent or longer-term biomarker monitoring might better predict outcomes.

Williams (2022) performed a post hoc analysis of the Multiple Sclerosis-Simvastatin Trial (MS-STAT) randomized controlled trial to determine whether baseline sNfL predicts cognitive decline in individuals with secondary progressive multiple sclerosis (SPMS) and whether it provides prognostic information beyond established MRI markers. The analysis included 110 participants (median age 50.7 years, median disease duration 20.7 years, median EDSS 6.0) with baseline sNfL measurements, MRI, and longitudinal neuropsychological testing over 24 months. The primary outcome was change in Wechsler Abbreviated Scale of Intelligence (WASI) Full Scale IQ. After adjustment for age, education, premorbid IQ, disability, treatment allocation, T2 lesion volume (T2LV), and regional brain volumes, each doubling of baseline sNfL was associated with a 0.010-point/month faster decline in WASI Full Scale IQ Z-score (95% CI, 0.003 to 0.017; p=0.008), equivalent to an estimated 0.24-point decline over 24 months. In contrast, MRI measures were primarily associated with baseline cognitive performance rather than future decline. Specifically, a 1 mL decrease in normalized transverse temporal gyrus volume was associated with a 0.362-point lower baseline WASI Full Scale IQ Z-score (95% CI, 0.026 to 0.698; p=0.035) but was not associated with subsequent cognitive decline. Secondary analyses demonstrated similar findings across multiple cognitive domains, with higher baseline sNfL associated with faster decline in verbal IQ, graded naming, and visuospatial performance, while lower cortical and deep gray matter volumes were more consistently associated with poorer baseline cognition. These findings suggest that sNfL may serve as an independent prognostic biomarker for future cognitive decline in SPMS beyond conventional MRI measures. Limitations include a secondary, post hoc analysis of a phase 2 clinical trial that was not originally designed or powered to evaluate serum biomarkers as predictors of cognitive decline. The analysis included only 110 participants, reducing statistical power and increasing the potential for type II error, particularly for secondary cognitive outcomes. Participants were limited to individuals with secondary progressive MS, restricting generalizability to relapsing-remitting MS (RRMS) or primary progressive MS. Follow-up was limited to 24 months, which may not fully capture the long-term trajectory of cognitive decline. The authors also noted that they were unable to calculate age- and body mass index-adjusted sNfL Z-scores, which have since been recommended for improved interpretation of NfL measurements. Additionally, longitudinal changes in sNfL and MRI measures were not independently associated with cognitive decline, and the study could not determine whether modifying sNfL levels would alter long-term cognitive outcomes.

Novakova (2017) evaluated whether sNfL could be used as a blood-based biomarker to monitor disease activity and treatment response in MS. The study analyzed 521 paired serum and cerebrospinal fluid (CSF) samples from 373 participants, including 286 individuals with MS, 45 with other neurologic conditions, and 42 healthy controls. Among those with MS, 204 had RRMS and 82 had progressive MS. sNfL was measured using an ultrasensitive single-molecule array (Simoa) assay, while CSF NfL was measured using an enzyme-linked immunosorbent assay (ELISA). Serum and CSF NfL were moderately to strongly correlated in MS (r=0.620; 95% CI, 0.558-0.675; p<0.001), supporting sNfL as a less invasive surrogate for CSF NfL. sNfL concentrations were significantly higher in RRMS (median 16.9 ng/L) and progressive MS (median 23.0 ng/L) compared with healthy controls (median 10.5 ng/L; p<0.001 for both). sNfL was also higher during relapse (19.1 ng/L) than remission (17.2 ng/L; p=0.043) and higher with gadolinium-enhancing MRI lesions (22.8 ng/L) than without such lesions (16.8 ng/L; p<0.001). sNfL increased with the number of gadolinium-enhancing lesions, where levels rose from 16.8 ng/L with no lesions to 31.9 ng/L with 2 or ≥3 lesions. Treatment with DMT reduced sNfL from 18.6 ng/L to 15.7 ng/L overall (p<0.001). In untreated individuals who initiated DMT, sNfL decreased from 22.7 to 20.2 ng/L (p=0.002), while those escalating to more effective DMT decreased from 17.7 to 12.4 ng/L (p<0.001). Individuals who remained untreated or changed to similarly effective DMTs had stable sNfL levels. Limitations include that the study used a real-world observational cohort, not a randomized trial, so treatment effects on NfL may be influenced by treatment selection, baseline disease activity, and other confounding factors. The follow-up subgroup was relatively heterogeneous, including untreated individuals, treatment initiators, DMT escalators, and individuals switching between therapies of similar efficacy. MRI assessment was limited because disease activity was defined using relapses and gadolinium-enhancing lesions, while new or enlarging T2 lesion data were not available, which the authors stated likely affected the sensitivity of sNfL for detecting disease activity. sNfL is a nonspecific marker of axonal injury and does not identify the cause of injury, meaning elevated levels may reflect MS activity, other neurologic injury, spinal cord lesions, diffuse white matter injury, gray matter pathology, or unrelated neurodegenerative processes. The study also noted that long-term outcomes were not assessed, so it could not determine whether serial sNfL monitoring predicts stepwise T2 lesion accumulation, disability progression, or conversion to progressive disease. In addition, individuals with progressive MS were excluded from the receiver operating characteristic (ROC) analysis because elevated NfL in this population may reflect degenerative processes rather than inflammatory disease activity alone.

Chitnis (2023) conducted a multicenter clinical validation study of the Octave MSDA Test, an 18-protein serum-based assay designed to quantify disease activity in multiple sclerosis (MS). The assay incorporates biomarkers representing 4 biologic pathways (Immunomodulation, Neuroinflammation, Myelin Biology, and Neuroaxonal Integrity) to generate 4 Disease Pathway scores and an overall Disease Activity score ranging from 1.0 to 10.0. The study included 614 serum samples from individuals with MS obtained from 4 centers and divided into a training cohort (n=426) and an independent test cohort (n=188). The primary endpoint was the presence or absence of gadolinium-enhancing (Gd+) MRI lesions, with secondary endpoints including new/enlarging (N/E) T2 lesions and active compared stable disease status. In the independent test cohort, the multi-protein model achieved an area under the receiver operating characteristic (AUROC) of 0.781 for detecting Gd+ lesions, 0.750 for detecting N/E T2 lesions, and 0.768 for identifying active compared to stable disease. These results significantly outperformed the best-performing single biomarker, NfL, which achieved AUROCs of 0.726, 0.660, and 0.683, respectively (bootstrapped one-sided p<0.05 for all comparisons). At the predefined score thresholds, samples with moderate/high Disease Activity scores had 4.49-fold greater odds of having ≥1 Gd+ lesion than those with low scores, while samples with high Disease Activity scores had 20.99-fold greater odds of having ≥2 Gd+ lesions than those with low/moderate scores. For detecting ≥1 Gd+ lesion in the test cohort, the score demonstrated a sensitivity of 73.7%, specificity of 61.6%, negative predictive value (NPV) of 77.5%, and accuracy of 66.5%. For detecting ≥2 Gd+ lesions, accuracy increased to 89.4% with an NPV of 95.1%. The investigators concluded that combining multiple biologically relevant biomarkers provided better discrimination of radiographic disease activity than NfL alone and may provide an objective adjunct to MRI and clinical assessment. Limitations include that the study was funded by Octave Bioscience, the manufacturer of the MSDA Test, and several authors were current or former employees of the company or reported financial relationships with Octave, creating the potential for conflict of interest. The manuscript was published as a medRxiv preprint and had not undergone peer review at the time of posting. The study population was intentionally enriched for Gd+ lesions, resulting in a higher prevalence of active disease than is typically encountered in routine clinical practice, which may inflate measures such as positive predictive value, negative predictive value, and overall accuracy. The model was developed and validated using a training/test split within the same overall dataset, rather than an entirely external validation cohort, which may overestimate model performance. Although samples were obtained from four centers, differences in sample collection and processing could introduce variability despite quality control procedures. The assay was validated primarily against MRI lesion activity and short-term clinical disease status; it was not validated for predicting long-term disability progression, relapse risk, or treatment response, and these analyses were specifically noted as areas for future investigation. The authors acknowledged that the test is intended to complement rather than replace MRI and clinical assessment and emphasized the need for larger real-world studies to further establish clinical utility.

Okuda (2026) conducted a retrospective observational study evaluating whether the Octave MSDA Test, an 18-protein blood-based assay, could help distinguish true MS relapse from pseudoexacerbation by comparing MSDA results with MRI findings. The study included 66 individuals with relapsing-remitting MS with 33 evaluated for acute neurologic symptoms concerning for relapse and 33 undergoing routine disease surveillance. Among symptomatic individuals, MSDA testing was performed before MRI in 79% of cases. Gadolinium-enhancing MRI lesions were identified in 10 symptomatic individuals and 6 asymptomatic individuals undergoing surveillance. A higher MSDA Disease Activity Score was strongly associated with gadolinium enhancement (p<0.0001). A DAS threshold of approximately ≥ 6.0 was reported as highly sensitive and specific for detecting gadolinium-enhancing lesions. Several proteins differed significantly in those with enhancing lesions, including higher myelin oligodendrocyte glycoprotein (MOG; p=0.013) and NfL; p<0.0001, and lower B-cell activating factor/B-cell activating factor (TNFSF13B; p<0.0001). The overall MSDA score predicted MRI enhancement better than NfL alone, with an AUC of 98.06% compared to 82.75% for NfL (p=0.0061). Limitations include the single-center retrospective design, small sample size, and limited generalizability. Social determinants of health, BMI, and renal function were not available, which may have affected timing of care or biomarker levels. MRI protocols and timing varied, and some elevated MSDA scores occurred without gadolinium enhancement, which could represent false positives or biologic activity not detected on MRI. The study also did not assess whether MSDA-guided decisions improve long-term outcomes.

Other blood-based biomarker tests for MS currently under investigation include but are not limited to: anti-KIR4 antibodies (Brickshawana, 2014; Brill, 2015; Navas-Madroñal, 2017), antiphospholipid antibodies (Koudriavtseva, 2014; Merashli, 2017), anti-myelin antibodies (Findling, 2014), osteopontin matrix protein biomarker (Agah, 2018), microRNAs (Regev, 2018), and NfL levels in plasma (Abdelhak, 2022; Hendricks, 2019; Sejbaek, 2019). However, these tests have not been proven to confirm a diagnosis of MS or alter disease management.

In 2025, Sanchez published the results of a retrospective chart review evaluating the clinical utility of the Octave MSDA test in routine MS care. The study analyzed 352 individuals’ charts across 14 United States (U.S.) clinics, assessing changes in treatment plans before and after MSDA test results. Most individuals had RRMS (87.5%), were female (78.1%), and had a mean age of 50 years. A total of 723 paired pre-/post-MSDA decision time points were reviewed. After receipt of MSDA results, clinicians changed the MS management plan in 19.4% of paired time points (95% CI, 13.3%-25.4%), exceeding predefined benchmark rates of 7% and 13% (p<0.001 and p=0.039, respectively). Decision changes were more common after the first MSDA result (23.4%) than after the second (16.1%) or third (10.3%) result. Clinicians reported that a single MSDA result influenced decision-making in 59.8% of time points, while longitudinal MSDA results influenced decisions in 69.2% of follow-up time points (95% CI, 60.2%-78.3%; p<0.001). Most MSDA scores were low (53.5%), followed by moderate (37.2%) and high (9.3%), and mean MSDA scores differed significantly across post-test management decisions (p<0.001), suggesting the test may influence decisions to continue, initiate, switch, or discontinue therapy. However, the study did not assess whether these management changes led to improved health outcomes such as reduced relapse rates, delayed disability progression, or enhanced quality of life. Limitations include a retrospective real-world utility study conducted only at sites already using MSDA, which may limit generalizability and introduce selection bias. Clinician-reported assessments may be affected by recall bias, although chart review was used to capture objective pre-test and post-test decisions. The study did not evaluate long-term outcomes, relapse rates, MRI outcomes, disability progression, or whether MSDA-guided decisions improved clinical results. Use of the test varied by clinician and practice setting, and several authors were current or former employees of Octave Bioscience, the test developer. The authors acknowledged that long-term prospective studies are needed to evaluate disease-modifying outcomes.

While the Octave MSDA Test demonstrates analytical and clinical validity and is associated with treatment decision changes, current evidence does not establish that its use results in improved health outcomes. No prospective studies have demonstrated reductions in relapse rates, delayed progression, or functional improvements attributable to the test.

Fung (2025) conducted a prospective biomarker analysis within the DOT-MS (Discontinuation of Therapy in Multiple Sclerosis) randomized clinical trial to evaluate whether sNfL and GFAP could detect disease reactivation following discontinuation of first-line DMT in individuals with stable relapsing-onset MS. Participants with long-term stable disease were randomized to either continue or discontinue first-line DMT and underwent serial clinical assessments, MRI, and serum biomarker measurements at baseline and at 3, 6, 12, 18, and 24 months. The primary biomarker outcome was the ability of changes in sNfL and GFAP to identify "significant" disease activity, defined as a clinical relapse, ≥ 3 new MRI lesions, or ≥ 2 contrast-enhancing lesions. Among 89 participants, 45 (50.5%) discontinued DMT. During follow-up, 8 participants (all in the discontinuation group) developed significant disease activity. Disease reactivation was associated with an increase in sNfL (odds ratio [OR], 1.13, 95% CI 1.03-1.33; p=0.04) and showed a trend with age- and body mass index-adjusted NfL z-scores (OR, 2.17, 95% CI 0.98-5.22; p=0.06), whereas GFAP was not associated with disease activity (p=0.52). The change in sNfL (delta NfL) demonstrated the strongest discriminatory ability for detecting significant disease activity, with an area under the receiver operating characteristic curve (AUC) of 0.88 (95% CI 0.76-0.99). A 46.4% increase in sNfL was identified as the optimal threshold for detecting disease reactivation, yielding 57% sensitivity and 96% specificity. These findings suggest that rising sNfL concentrations may serve as an early marker of inflammatory disease recurrence after DMT withdrawal, whereas GFAP did not demonstrate similar utility. Limitations include a small sample size, with only 89 participants and 8 disease activity events, limiting statistical power and precision of the estimated diagnostic performance. Since all disease activity occurred in the DMT discontinuation group, the findings may not be generalizable to broader MS populations or other treatment settings. The calculated biomarker cut-off was derived from this cohort and requires external validation before clinical implementation. Although sNfL demonstrated excellent overall discrimination (AUC 0.88), the proposed threshold had modest sensitivity (57%), indicating that many individuals with disease reactivation would not be detected by this cut-off alone. Additionally, biomarker measurements were performed at scheduled visits rather than continuously, and longer follow-up is needed to determine whether serial biomarker monitoring improves long-term clinical outcomes.

Corsten (2026) conducted a prospective cohort study to evaluate whether sNfL and GFAP, individually or in combination, could predict the development of MS after a CIS. The study enrolled 221 adults with CIS within 6 months of symptom onset. Baseline serum NfL and GFAP concentrations were measured and converted to age-adjusted z-scores. The investigators also assessed established MS risk factors, including HLA-DRB1*15:01 carrier status, weighted genetic risk scores, and anti-Epstein-Barr virus nuclear antigen-1 (anti-EBNA1) IgG antibodies. Participants were followed prospectively, and the primary outcome was diagnosis of McDonald 2024 MS. During follow-up, 162 of 221 participants (73.3%) were diagnosed with MS. In separate Cox proportional hazards models, higher baseline NfL (HR, 1.36; 95% CI, 1.12-1.66; p=0.002) and GFAP (HR, 1.12; 95% CI, 1.02-1.42; p=0.01) were each associated with earlier MS diagnosis. However, when both biomarkers were included in the same model, only NfL remained an independent predictor (HR, 1.30; 95% CI, 1.02-1.60; p=0.01), indicating that GFAP provided little additional prognostic information. Time-dependent ROC analyses similarly demonstrated that NfL alone performed as well as the combined NfL/GFAP model, with AUCs of approximately 0.81 at 1 year, 0.80 at 3 years, and 0.76 at 5 years. Adding HLA-DRB1*15:01 modestly improved predictive performance, whereas inclusion of GFAP or anti-EBNA1 IgG did not meaningfully enhance discrimination. Elevated NfL was also associated with markers of acute inflammatory disease activity, including greater baseline T2 lesion burden and gadolinium-enhancing MRI lesions. Limitations include that the study was conducted at a single MS center, which may limit generalizability. Biomarkers were measured only at baseline, so the study could not evaluate whether serial measurements improve prediction. Most participants had MRI findings suggestive of demyelinating disease, resulting in a relatively high conversion rate to MS (73%), which may not reflect all CIS populations. The proposed NfL z-score threshold (2.26) for risk stratification was derived from this cohort and requires external validation before routine clinical use. Additionally, the study focused on prediction of early MS diagnosis rather than long-term disability progression or treatment response.

Tallantyre (2026) published a narrative review describing the current state of blood-based biomarkers for MS and their potential role in improving diagnosis, prognosis, disease monitoring, treatment selection, and understanding of disease biology. The review highlights advances in ultrasensitive assay technologies and multi-omics approaches that have enabled reliable measurement of biomarkers in blood rather than CSF. Among currently available biomarkers, sNfL has the strongest evidence and is considered the closest to routine clinical implementation. Elevated NfL levels have consistently been associated with acute inflammatory disease activity, MRI lesion burden, relapse risk, brain atrophy, disability progression, and treatment response across multiple studies. GFAP was also identified as a promising biomarker, particularly for progressive MS, where higher concentrations correlate with disease progression and chronic neurodegeneration. Additional candidate biomarkers discussed include markers of immune activation, myelin injury, axonal damage, complement activation, B-cell biology, Epstein-Barr virus-related immunity, and proteomic, metabolomic, and microRNA-based signatures. The authors emphasize that combinations of biomarkers integrated with clinical and MRI findings will likely provide greater predictive value than individual biomarkers alone. Limitations include that this is a narrative review rather than a systematic review or meta-analysis, and therefore does not include pooled statistical analyses or formal assessment of study quality. The conclusions are based on previously published literature, which varies considerably in study design, assay methodology, populations, and outcome measures. Many candidate biomarkers remain investigational, with limited external validation and no universally accepted reference ranges or clinical thresholds. The authors note that NfL is sensitive but not disease-specific, as elevated levels occur in numerous neurologic disorders, limiting its diagnostic specificity for MS. Significant barriers remain before widespread clinical implementation, including assay standardization across laboratories, establishment of age-, sex-, and body mass index-adjusted reference values, regulatory approval, and consensus regarding how biomarkers should complement or modify existing MRI-based monitoring strategies.

In 2013, Ouallet reported on the Multiple Sclerosis Think Tank (GRESEP) which is a French expert working group composed of hospital-based neurologists specializing in MS. GRESEP was established to address clinical questions encountered in routine MS practice that were either only partially addressed during the 2001 French Federation of Neurology Consensus Conference or had evolved with emerging evidence. The group focused its work on four key topics: cognition in MS, early disease management, the role of serum testing in the diagnosis of MS, and management of MS exacerbations. They addressed the clinical utility of serum biomarkers in the diagnosis and differential diagnosis of MS. The first recommendation had strong professional agreement and stated, “there is currently no useful biological blood test for the positive diagnosis of multiple sclerosis”.

In 2014, the Advisory Committee on Clinical Trials in MS, the US National Multiple Sclerosis Society, the European Committee for Treatment and Research in MS, and other experts (the MS Phenotype Group) published a re-examination of MS clinical course descriptions (Lublin, 2014). The authors indicated that there may be markers of disease activity (other than clinical exacerbations or MRI-detected lesions) but “there is insufficient evidence for including them at this time.” Additionally, the committee stated:

To date, there are no clear clinical, imaging, immunologic, or pathologic criteria to determine the transition point when RRMS converts to SPMS; the transition is usually gradual. This has limited our ability to study the imaging and biomarker characteristics that may distinguish this course.

Other Relevant Information

No FDA labeled indications have been identified for the tests that conduct analysis of blood based biomarkers for MS. Two blood-based tests measuring NfL, the Quanterix Simoa^® NfL Plasma Test (Quanterix Corporation, Billerica, MA) and the Roche Elecsys^® NfL Test (Roche Diagnostics International Ltd., Basel, Switzerland), have been granted FDA Breakthrough Device Designation for use in MS, but neither has received FDA approval. The Octave Multiple Sclerosis Disease Activity (MSDA) test is available in the U.S. as laboratory-developed test (LDT), without FDA review. Neither Centers for Medicare and Medicaid Services (CMS) National Coverage Determination (NCD) nor Local Coverage Determinations (LCD) addressing blood biomarkers for MS were identified. No nationally recognized clinical practice guidelines currently address blood biomarkers for MS.

Multiple sclerosis (MS) is an autoimmune disease of the central nervous system (CNS). During the MS disease process, inflammation of nervous tissue causes the loss of myelin, a fatty material that acts as a protective insulation for the nerve fibers in the brain and spinal cord. This demyelination leaves multiple areas of hard, scarred tissue (plaques) along the covering of the nerve cells. Another characteristic of MS is the destruction of axons, which are the long filaments that carry electric impulses away from a nerve cell. Demyelination and axon destruction disrupt the ability of the nerves to conduct electrical impulses to and from the brain and produce various symptoms. Common symptoms of the disease include fatigue, numbness, coordination and balance problems, bowel and bladder dysfunction, emotional and cognitive changes, spasticity, vision problems, dizziness, sexual dysfunction, and pain. Classifications of MS are relapsing-remitting multiple sclerosis (RRMS), primary progressive multiple sclerosis (PPMS), progressive relapsing multiple sclerosis (PRMS), and secondary progressive multiple sclerosis (SPMS). Most individuals with MS have a relapsing course, and their first attack may present as a clinically isolated syndrome (CIS). A CIS is a single demyelinating episode with consistent magnetic resonance imaging (MRI) findings (indicating inflammation/demyelination in one site in the CNS). Individuals with CIS are at high risk for developing clinically definite MS.

Clinically isolated syndrome (CIS): A first neurologic event that is suggestive of demyelination, accompanied by multiple, clinically “silent” (asymptomatic) lesions on magnetic resonance imaging (MRI) that are typical of multiple sclerosis (MS). Individuals with this syndrome are at high risk for developing clinically definite MS.

Relapsing-remitting multiple sclerosis (RRMS): An inflammatory demyelinating disease of the central nervous system (CNS) in which individuals experience intermittent relapses (also called attacks or exacerbations) resulting from new inflammatory lesions in the brain, spinal cord, or optic nerves. These relapses are followed by remissions, during which symptoms improve either partially or completely. Between relapses, there is generally no sustained progression of neurologic disability, although residual deficits from prior attacks may persist.

The following codes for treatments and procedures applicable to this document are included below for informational purposes. Inclusion or exclusion of a procedure, diagnosis or device code(s) does not constitute or imply member coverage or provider reimbursement policy. Please refer to the member's contract benefits in effect at the time of service to determine coverage or non-coverage of these services as it applies to an individual member.

When services are Investigational and Not Medically Necessary:
When the code describes a procedure indicated in the Position Statement section as investigational and not medically necessary.

Peer Reviewed Publications:

1. Abdelhak A, Cordano C, Boscardin WJ, et al. Plasma neurofilament light chain levels suggest neuroaxonal stability following therapeutic remyelination in people with multiple sclerosis. J Neurol Neurosurg Psychiatry. 2022; 93:972-977.
2. Agah E, Zardoui A, Saghazadeh A, et al. Osteopontin (OPN) as a CSF and blood biomarker for multiple sclerosis: a systematic review and meta-analysis. PLoS One. 2018; 13(1):e0190252.
3. Aktas O, Renner A, Huss A, et al. Serum neurofilament light chain: no clear relation to cognition and neuropsychiatric symptoms in stable MS. Neurol Neuroimmunol Neuroinflamm. 2020; 7(6):e885.
4. Benkert P, Meier S, Schaedelin S, et al. Serum neurofilament light chain for individual prognostication of disease activity in people with multiple sclerosis: a retrospective modelling and validation study. Lancet Neurol. 2022; 21(3):246-257.
5. Brettschneider J, Jaskowski TD, Tumani H, et al. Serum anti-GAGA4 IgM antibodies differentiate relapsing remitting and secondary progressive multiple sclerosis from primary progressive multiple sclerosis and other neurological diseases. J Neuroimmunol. 2009; 217(1-2):95-101.
6. Brickshawana A, Hinson SR, Romero MF, et al. Investigation of the KIR4.1 potassium channel as a putative antigen in patients with multiple sclerosis: a comparative study. Lancet Neurol. 2014; 13(8):795-806.
7. Brill L, Goldberg L, Karni A, et al. Increased anti-KIR4.1 antibodies in multiple sclerosis: could it be a marker of disease relapse? Mult Scler. 2015; 21(5):572-579.
8. Cai L, Huang J. Neurofilament light chain as a biological marker for multiple sclerosis: a meta-analysis study. Neuropsychiatr Dis Treat. 2018; 14:2241-2254.
9. Calabresi PA, Arnold DL, Sangurdekar D, et al. Temporal profile of serum neurofilament light in multiple sclerosis: implications for patient monitoring. Mult Scler. 2021; 27(10):1497-1505.
10. Chitnis T, Foley J, Ionete C, et al. Clinical validation of a multi-protein, serum-based assay for disease activity assessments in multiple sclerosis. Clinical immunology. 2023; 253:109688.
11. Corsten CEA, Geraedts VSA, Marques AM, et al. Serum neurofilament light chain and glial fibrillary acidic protein predicting multiple sclerosis after clinically isolated syndrome. J Neurol. 2026; 273(2):128.
12. Findling O, Durot I, Weck A, et al. Antimyelin antibodies as predictors of disability after clinically isolated syndrome. Int J Neurosci. 2014; 124(8):567-572.
13. Freedman MS, Laks J, Dotan N, et al. Anti-alpha-glucose-based glycan IgM antibodies predict relapse activity in multiple sclerosis after the first neurological event. Mult Scler. 2009; 15(4):422-430.
14. Freedman MS, Metzig C, Kappos L, et al. Predictive nature of IgM anti-α-glucose serum biomarker for relapse activity and EDSS progression in CIS patients: a BENEFIT study analysis. Mult Scler. 2012; 18(7):966-973.
15. Fung WH, Wessels MHJ, Coerver EME, et al. Reliability of serum neurofilament light and glial fibrillary acidic protein for detecting disease activity upon discontinuation of first-line disease-modifying therapy in stable multiple sclerosis (DOT-MS). J Neurol. 2025; 272(8):530.
16. Hanninen K, Jaaskelainen O, Herukka SK, Soilu-Hanninen M. Vitamin D supplementation and serum neurofilament light chain in interferon-beta-1b-treated MS patients. Brain Behav. 2020; 10(9):e01772.
17. Hendricks R, Baker D, Brumm J, Det al. Establishment of neurofilament light chain Simoa assay in cerebrospinal fluid and blood. Bioanalysis. 2019; 11(15):1405-1418.
18. Koudriavtseva T, D'Agosto G, Mandoj C, et al. High frequency of antiphospholipid antibodies in relapse of multiple sclerosis: a possible indicator of inflammatory-thrombotic processes. Neurol Sci. 2014; 35(11):1737-1741.
19. Merashli M, Alves JD, Gentile F, Ames PRJ. Relevance of antiphospholipid antibodies in multiple sclerosis: a systematic review and meta analysis. Semin Arthritis Rheum. 2017; 46(6):810-818.
20. Navas-Madroñal M, Valero-Mut A, Martínez-Zapata MJ, et al. Absence of antibodies against KIR4.1 in multiple sclerosis: a three-technique approach and systematic review. PLoS One. 2017; 12(4): e0175538.
21. Novakova L, Zetterberg H, Sundström P, et al. Monitoring disease activity in multiple sclerosis using serum neurofilament light protein. Neurology. 2017; 89(22):2230-2237.
22. Okuda DT, Moog TM, McCreary MC, et al. Utility of multi-analyte protein assay to distinguish multiple sclerosis clinical relapse from pseudoexacerbation. Neurol Ther. 2026 Jun 27; [Epub ahead of print]. PMID: 42364069.
23. Ouallet JC, Bodiguel E, Bensa C, et al; Groupe de Re´flexion sur la Scle´rose en Plaques: GRESE. Recommendations for useful serum testing with suspected multiple sclerosis. Rev Neurol (Paris). 2013; 169(1):37-46.
24. Regev K, Healy BC, Paul A, et al. Identification of MS-specific serum miRNAs in an international multicenter study. Neurol Neuroimmunol Neuroinflamm. 2018; 5(5):e491.
25. Rosso M, Gonzalez CT, Healy BC, et al. Temporal association of sNfL and gad-enhancing lesions in multiple sclerosis. Ann Clin Transl Neurol. 2020; 7(6):945-955.
26. Sanchez A, Sheng E, Eagleman S, et al. Real-world clinical utility of a multi-protein, blood-based biomarker assay for disease activity assessments in multiple sclerosis. Mult Scler J Exp Transl Clin. 2025; 11(2):20552173251331030.
27. Schwarz M, Spector L, Gortler M, et al. Serum anti-Glc(alpha1,4)Glc(alpha) antibodies as a biomarker for relapsing-remitting multiple sclerosis. J Neurol Sci. 2006; 244(1-2):59-68.
28. Seiberl M, Feige J, Hilpold P, et al. Serum neurofilament light chain as biomarker for cladribine-treated multiple sclerosis patients in a real-world setting. Int J Mol Sci. 2023; 24(4):4067.
29. Sejbaek T, Nielsen HH, Penner N, et al. Dimethyl fumarate decreases neurofilament light chain in CSF and blood of treatment naïve relapsing MS patients. J Neurol Neurosurg Psychiatry. 2019; 90(12):1324-1330.
30. Tallantyre EC, Apap Mangion S, Teunissen CE. A framework for blood biomarker discovery in MS. Mult Scler. 2026; 32(7):661-669.
31. Williams T, Tur C, Eshaghi A, et al. Serum neurofilament light and MRI predictors of cognitive decline in patients with secondary progressive multiple sclerosis: analysis from the MS-STAT randomised controlled trial. Mult Scler. 2022; 28(12):1913-1926.

Government Agency, Medical Society, and Other Authoritative Publications:

1. Lublin FD, Reingold SC, Cohen JA, et al. Defining the clinical course of multiple sclerosis: the 2013 revisions. Neurology. 2014; 83(3):278-286.

1. National Multiple Sclerosis Society. About MS. Last updated August 2025. Available at: https://www.nationalmssociety.org/understanding-ms/what-is-ms. Accessed on July 30, 2026.

Glial fibrillary acidic protein (GFAP)
Glycan-Based Test
Multiple Sclerosis Biomarkers
Neurofilament light chain
Octave MSDA test
Serum Biomarker Tests for Multiple Sclerosis
Serum Neurofilament Light Chain

The use of specific product names is illustrative only. It is not intended to be a recommendation of one product over another, and is not intended to represent a complete listing of all products available.

Federal and State law, as well as contract language, including definitions and specific contract provisions/exclusions, take precedence over Medical Policy and must be considered first in determining eligibility for coverage. The member’s contract benefits in effect on the date that services are rendered must be used. Medical Policy, which addresses medical efficacy, should be considered before utilizing medical opinion in adjudication. Medical technology is constantly evolving, and we reserve the right to review and update Medical Policy periodically.

No part of this publication may be reproduced, stored in a retrieval system or transmitted, in any form or by any means, electronic, mechanical, photocopying, or otherwise, without permission from the health plan.

© CPT Only – American Medical Association